Vol.:(0123456789)1 3

Cancer Immunology, Immunotherapy (2018) 67:1897–1910 https://doi.org/10.1007/s00262-018-2157-5

FOCUSSED RESEARCH REVIEW

Tumor lysate-based vaccines: on the road to immunotherapy for gallbladder cancer

Daniel Rojas‑Sepúlveda1,2,3 · Andrés Tittarelli1,2  · María Alejandra Gleisner1,2 · Ignacio Ávalos1,2 · Cristián Pereda1,2 · Iván Gallegos5 · Fermín Eduardo González2,4 · Mercedes Natalia López1,2 · Jean Michel Butte6 · Juan Carlos Roa7,8 · Paula Fluxá1,2 · Flavio Salazar‑Onfray1,2

Received: 1 September 2017 / Accepted: 26 March 2018 / Published online: 29 March 2018 © The Author(s) 2018

AbstractImmunotherapy based on checkpoint blockers has proven survival benefits in patients with melanoma and other malignan-cies. Nevertheless, a significant proportion of treated patients remains refractory, suggesting that in combination with active immunizations, such as cancer vaccines, they could be helpful to improve response rates. During the last decade, we have used dendritic cell (DC) based vaccines where DCs loaded with an allogeneic heat-conditioned melanoma cell lysate were tested in a series of clinical trials. In these studies, 60% of stage IV melanoma DC-treated patients showed immunological responses correlating with improved survival. Further studies showed that an essential part of the clinical efficacy was associ-ated with the use of conditioned lysates. Gallbladder cancer (GBC) is a high-incidence malignancy in South America. Here, we evaluated the feasibility of producing effective DCs using heat-conditioned cell lysates derived from gallbladder cancer cell lines (GBCCL). By characterizing nine different GBCCLs and several fresh tumor tissues, we found that they expressed some tumor-associated antigens such as CEA, MUC-1, CA19-9, Erb2, Survivin, and several carcinoembryonic antigens. Moreover, heat-shock treatment of GBCCLs induced calreticulin translocation and release of HMGB1 and ATP, both known to act as danger signals. Monocytes stimulated with combinations of conditioned lysates exhibited a potent increase of DC-maturation markers. Furthermore, conditioned lysate-matured DCs were capable of strongly inducing CD4+ and CD8+ T cell activation, in both allogeneic and autologous cell co-cultures. Finally, in vitro stimulated CD8+ T cells recognize HLA-matched GBCCLs. In summary, GBC cell lysate-loaded DCs may be considered for future immunotherapy approaches.

Keywords Melanoma · Gallbladder cancer · Dendritic cells · Tumor lysates · Immunotherapy · CITIM 2017

AbbreviationsAM Activated monocytesBAGE B melanoma antigen

CA19-9 Cancer antigen 19-9CCR7 C-C chemokine receptor type 7CEA Carcinoembryonic antigenCXCR C-X-C motif chemokine receptorDAMPs Damage associated molecular patternsDC Dendritic celleCRT Extracellular (or exo) calreticulinFBS Fetal bovine serumGAGE G antigenGBC Gallbladder cancerGBCCL Gallbladder cancer cell lineHLA Human leukocyte antigenHMGB1 High mobility group box-1IFN-γ Interferon gammaiMFI Integrated mean fluorescence intensityMAGE Melanoma-associated antigenM2-DCs Dendritic cells matured with M2 lysateM3-DCs Dendritic cells matured with M3 lysate

Prior to presentation at CITIM 2017 in Prague, part of this work was also presented at the International Congress of Immunology (ICI), 21–26 August 2016, Melbourne, Australia. An abstract was published in [1]. Following presentation at CITIM 2017, part of this work was also presented at IMMUNOLOGY 2017, 12–16 May 2017, Washington, USA. An abstract was published in [2].

This paper is a Focussed Research Review based on a presentation given at the Fifth International Conference on Cancer Immunotherapy and Immunomonitoring (CITIM 2017), held in Prague, Czech Republic, 24th–27th April 2017. It is part of a series of Focussed Research Reviews and meeting report in Cancer Immunology, Immunotherapy.

* Flavio Salazar-Onfray fsalazar@u.uchile.cl

Extended author information available on the last page of the article

1898 Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

M5-DCs Dendritic cells matured with M5 lysateM8-DCs Dendritic cells matured with M8 lysatemAbs Monoclonal antibodiesMHC Major histocompatibility complexMUC-1 Mucin-1PBMC Peripheral blood mononuclear cellsrhGM-CSF Recombinant human granulocyte–mac-

rophage colony-stimulating factorrhIL Recombinant human interleukinTAAs Tumor associated antigensTh1 Type 1 T helper cellsTh2 Type 2 T helper cellsTNF-α Tumor necrosis factorαTRIMEL-DCs Dendritic cells matured with TRIMEL

Introduction

The recent use of immune-checkpoint blocker antibod-ies has demonstrated durable clinical benefits in patients with melanoma, lung cancer and other solid tumors [3–9]. Despite this relevant clinical performance, a high percentage of treated patients remains refractory, strongly suggesting that the combination with active immunizations may be use-ful to improve the response rates of those patients. In this context, cancer vaccines, particularly dendritic cell (DC)-based vaccines, can be used as complementary treatments in cancer patients. Optimal delivery of a wide-ranging pool of tumor-associated antigens (TAAs) and the use of adequate adjuvants are shown to be crucial for vaccine success [10]. During the last decade, we have been able to produce thera-peutic DCs using an allogeneic heat-conditioned melanoma cell lysate named TRIMEL. Sixty percent of advanced mela-noma patients treated with these DCs showed a delayed type hypersensitivity reaction against TRIMEL, which correlated with a threefold prolonged survival [11]. This strategy pro-vides a reproducible pool of almost all the potential mela-noma-associated antigens, suitable for use in a wide range of patients independent of their major histocompatibility complex (MHC) haplotypes or the availability of autolo-gous tumor tissue [12]. Moreover, we previously showed that TRIMEL contains some heat shock-induced damage-associated molecular patterns (DAMPs), such as high mobil-ity group box-1 (HMGB1) and calreticulin (eCRT), which mediate an optimal maturation, activation and antigen cross-presentation of the monocyte-derived DCs, and thus enable them to activate antigen-specific T cells [13]. However, the development of an optimal allogeneic tumor cell lysate preparation for different tumor types is crucial to expand the use of these approaches for different cancers.

Gallbladder cancer (GBC) is the most common cancer of the biliary tree. Although GBC is infrequent in developed countries [14], in South America and particularly in Chile,

this tumor constitutes a major health problem [14–17]. The underlying causes for the high risk of GBC in these areas are unclear, but several important risk factors probably contrib-ute, including chronic inflammation caused by gallstones, high obesity rates and genetic susceptibility in women of indigenous Mapuche ancestry, in which the incidence increases to 27.3 cases per 100,000 [14, 16, 17].

Early detection and diagnosis of GBC is complicated because the clinical symptoms are manifested in advanced stages. The average survival time for patients with advanced, non-resectable GBC varies from 4 to 14 months [17, 18]. The most effective treatment for this type of cancer is surgi-cal removal of the primary tumor and areas of local exten-sion. Unfortunately, less than 10% of patients have resectable tumors, and nearly 50% of them present metastasis at the time of diagnosis [19]. Even with surgery, most of the GBC patients progress to a metastatic stage, highlighting the need for novel adjuvant therapies, such as immunotherapy.

The purpose of this study was to investigate the immu-nogenicity of several combinations of tumor lysates derived from different GBC cell lines (GBCCL) and their effect on monocyte differentiation and activation to DCs and their capacity to induce an in vitro T cell-mediated anti-GBC response. In this respect, a major requirement for the potential clinical effectiveness of GBC lysate-loaded DCs is to investigate the presence of shared TAAs in GBCCL and in fresh tumor tissues. Our results suggest that human DCs matured with specific GBCCL heat shock-conditioned lysates are capable of inducing specific T cells activation against this tumor and can be considered for the develop-ment of future immunotherapeutic approaches for GBC patients.

Materials and methods

Cell lines and cell lysates

GBCCL GBd1 (CVCL_H705), G415 (CVCL_8198), OCUG-1 (CVCL_3083), NOZ (CVCL_3079), TGBC-1TKB (CVCL_1769; hereafter 1TKB), TGBC-2TKB (CVCL_3339; hereaf ter 2TKB), TGBC-14TKB (CVCL_3340; hereafter 14TKB) and TGBC-24TKB (CVCL_1770; hereafter 24TKB) were provided by Juan Carlos Roa (Department of Pathology, Pontificia Univer-sidad Católica de Chile, Santiago, Chile). The GBCCL CAVE was established in our lab from a primary adenocar-cinoma GBC tumor sample from a Chilean patient. NOZ, GBd1 and G415 cells were grown in RPMI 1640 culture medium (Corning, NY, USA), whereas OCUG-1, 1TKB, 2TKB, 14TKB, 24TKB and CAVE were grown in DMEM culture medium (Corning, NY, USA). Culture media were supplemented with 10% fetal bovine serum (FBS), 10 U/

1899Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

mL penicillin and 10 mg/mL streptomycin (Corning, NY, USA). Cells were maintained at 37 °C under 5% CO2 and 95% relative humidity.

Cell lysates were produced as previously described [13]. Briefly, for individual GBCCL lysates, 4 × 106 cells/mL were heat shocked at 42 °C for 1 h, incubated for 2 h at 37 °C and then lysed. For GBCCL combined lysates, cells were mixed in equal amounts to achieve a final concentration of 4 × 106 cells/mL, and heat shocked as described before. The mixed cell lysates evaluated were made as follows: M1 (24TKB + GBd1 + G415); M 2 ( 2 T K B + 2 4 T K B + G B d 1 ) ; M 3 (1TKB + 2TKB + 24TKB); M4 (OCUG1 + GBd1 + G415); M5 (2TKB + G415 + OCUG1); M6 (NOZ + OCUG 1 + G415); M7 (1TKB + 14TKB + 24TKB); and M8 (24TKB + OCUG1 + G415).

Antibodies

Monoclonal antibodies (mAbs) against human carcinoem-bryonic antigen (CEA; clone COL-1), erbB2 (clone 3B5), and survivin (clone 8E2) were purchased from Thermo Fisher Scientific (Waltham, Massachusetts, USA). mAbs against human mucin-1 (MUC-1; clone HMFG1), cancer antigen 19-9 (CA19-9; clone SPM110) and calreticulin (clone FMC 75) were purchased from Abcam (Cambridge, USA). mAbs against human CD3 eFluor450 (clone SK7), human leukocyte antigen (HLA)-DR APC eFluor780 (clone LN3), CD83 PE Cy7 (clone HB15e), CD25 PerCP Cy5.5 (clone BC96), CD69 PE (clone FN50) and interleukin (IL)-4 PE Cy7 (clone 8D4-8) were purchased from eBioscience (San Diego, CA, USA). mAbs against human CD8 PE Cy7 (clone SK1), C-C chemokine receptor type 7 (CCR7) PE (clone G043H7), CD4 APC Cy7 (clone RPA-T4), tumor necrosis factor (TNF)-α PerCP (clone Mab11) and interferon (IFN)-γ AlexaFluor 647 (clone 4S.B3) were purchased from BioLegend (San Diego, CA, USA). Polyclonal goat anti-mouse IgG antibody was purchased from eBioscience. mAbs against human HLA-ABC (clone G46-2.6), CD80 BV421 (clone L307.4), CD86 BB515 (clone 2331), C-X-C motif chemokine receptor (CXCR)3 APC (clone 1C6/CXCR3) and CXCR4 APC (clone 12G5) were purchased from BD Pharmingen (San Diego, CA, USA).

Flow cytometry

The surface expression of MUC-1, erbB2, survivin, CA19-9, CEA, and eCRT was analyzed by flow cytometry. Intracel-lular staining was performed with the Foxp3/Transcription Factor Fixation/Permeabilization Concentrate and Diluent kit (eBioscience). Live/dead kit (Thermo Fisher) was used for live/dead cell discrimination. Flow cytometry was con-ducted on a FACSVerse flow cytometer (BD Biosciences)

and data analysis was performed using the FlowJo software (Tree Star, Inc., Ashland, OR, USA).

Reverse transcription polymerase chain reaction (RT‑PCR)

Total RNA was extracted from cells using TriPure reagent (Roche) and used to determine the expression and relative level of the Melanoma-associated antigen (MAGE), G anti-gen (GAGE) and B melanoma antigen (BAGE) in GBCCL. cDNAs were synthesized with M-MLV Reverse Tran-scriptase (Life Technologies). PCR was performed using cDNA template in the MasterCycler (Eppendorf), accord-ing to the manufacturer’s instructions. The sequences of the used primers are available under request.

Immunohistochemistry

Sections of 3 µm thickness from paraffin-embedded GBC tissues were mounted on slides, rehydrated and antigen retrieval was performed by heat in Tris–EDTA pH 9.0 or citrate buffer pH 6.0 depending on the Ab used. Primary Abs were used according to manufacturer’s instructions (CEA dilution 1:200, clone COL-1, Thermo Scientific; MUC-1 dilution 1:200, clone HMFG1, Abcam; erbB2 dilu-tion 1:200, clone 3B5, Thermo Scientific; CA19-9 dilution 1:50, clone SPM110, Abcam; and survivin dilution 1:50, clone 8E2, Thermo Scientific). The slides were incubated with primary Abs in a moist chamber overnight at 4 °C. After incubation with primary Abs, slides were washed with TBS before incubation with labeled secondary Abs for 1 h at 4 °C. Sections were subsequently incubated with ABC solu-tion for 30 min (ABC Vectastain Kit Elite PK6200, Vector Laboratories), washed with three changes of TBS, incubated with Dako-Chromogen solution and washed with deionized water. Background staining was performed with Mayer’s hematoxylin, sections were dehydrated through ascending alcohols to xylene and mounted. Negative control slides omitting the primary Ab were included in all batches. An expert pathologist evaluated the expressions of CEA, MUC-1, c-erbB2, CA19-9 and survivin in GBC tissues.

Enzyme‑linked immunosorbent assay (ELISA)

The concentration of HMGB1 in 100 µL of supernatants from control and heat shocked GBCCL (4 × 106 cells/mL) were measured by ELISA using a specific HMGB1 ELISA kit according to the manufacturer’s instructions (Cloud-Clone Corp.). 450 nm optical densities were measured in a Sunrise absorbance reader (Tecan).

1900 Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

ATP determination

The concentration of ATP in supernatants from control and heat shocked GBCCL (4 × 106 cells/mL) was measured by the Luciferase-Based ATP Determination Kit (Life Tech-nologies) according to the manufacturer’s instructions. Luminescence was measured in a TopCount luminescence counter (PerkinElmer).

DC generation

Adherent monocytes isolated from peripheral blood mono-nuclear cells (PBMC) of healthy donors from the Centro Metropolitano de Sangre y Tejidos, Hospital Metropoli-tano (Santiago, Chile) were cultured in serum-free AIM-V medium (Invitrogen) for 22 h with 500 U/mL recombinant human IL-4 (rhIL-4; US-Biological) and 800 U/mL recom-binant human granulocyte–macrophage colony-stimulating factor (rhGM-CSF; Sheering Plough) and then stimulated for 24 h with 100 µg/mL of GBCCL lysates, TRIMEL (TRI-MEL-DCs) or with medium [activated monocytes (AM)] as previously described [20].

DC/T cell co‑cultures

For allogeneic cell co-cultures, CD3+ T cells from healthy donors were sorted with a FACSAria II sorter (BD Bio-sciences) and co-cultured for 5 days with TRIMEL-DCs or DCs matured with GBCCL lysates at a 20:1 ratio in RPMI 1640 medium supplemented with 10% FBS and 150 UI/mL rhIL-2 (Proleukin). For autologous co-cultures, sorted CD3+ T cells from HLA-A2+ healthy donors were co-cul-tured with AM, TRIMEL-DCs or DCs matured with the M2 lysate (M2-DCs) for 14 days at a 10:1 ratio in RPMI 1640 medium supplemented with 10% FBS and 150 UI/mL rhIL-2. T cells were re-stimulated at day 7 with freshly prepared DCs maintaining the cell:cell ratio. Surface expression of CD25, CD69, CXCR3 and CXCR4 was analyzed in CD4+ and CD8+ T cells by flow cytometry. For intracellular IFN-γ, TNF-α and IL-4 staining, 1 × 106 T cells were cultured for 4 h at 37 °C in RPMI 1640 medium with 10% FBS contain-ing 1 µg/mL ionomycin, 0.15 µM phorbol myristate acetate (PMA), and 3 µg/mL brefeldin A. T cell proliferation was studied using carboxyfluorescein succinimidyl ester (CFSE) dilution analysis.

IFN‑γ ELISpot

Autologous CD8+ T cells activated with AM, TRIMEL-DCs or M2-DCs were sorted and co-cultured with 1 × 104 target cells: HLA-A2+ GBCCL (GBd1, TGBC-2TKB, CAVE), HLA-A2+ melanoma cell line (Mel1) or K562 for 16 h at different effector/target ratios. IFN-γ release was tested by

an ELISpot assay according to the manufacturer’s instruc-tions (ELISPOT Ready-SET-Go, eBioscience) as previously described [20].

Statistical analysis

Statistical analysis was achieved using GraphPad Prism software version 6.0 (GraphPad Software, San Diego, CA, USA). Student’s t test was used to determine differ-ences between treatments and results are presented as mean ± standard deviation (SD). p values < 0.05 were con-sidered significant.

Results

GBCCL express relevant tumor‑associated antigens present in GBC tissues

To select a GBCCL suitable for the production of cell lysates as a source of multiple tumor antigens, the levels of expres-sion of 10 of the most common and relevant TAAs (sur-vivin, MUC-1, CEA, erbB2, CA19-9, MAGE-1, MAGE- 2, MAGE-3, GAGE-1/2 and BAGE) were determined in eight publicly available GBCCL (GBd1, G415, OCUG-1, NOZ, 1TKB, 2TKB, 14TKB and 24TKB) and in one GBCCL established in our lab (CAVE). The protein levels of sur-vivin, MUC-1, CEA, erbB2 and CA19-9 were determined by flow cytometry, whereas the expression of MAGEs, GAGEs and BAGE was evaluated at the RNA level by RT-PCR. The nine GBCCL showed diverse levels and patterns of antigen expression and none of them expressed all ten antigens, but all expressed at least two of them (Fig. 1a–c). The expression of erbB2 was detected in all the cell lines analyzed, whereas the 2TKB cells expressed only the antigens GAGE1/2 and BAGE. The cell lines with the broader pattern of antigen expression were 2TKB and 1TKB, which express 8 and 7 of the 10 antigens, respectively (Fig. 1c). Additionally, sur-vivin, MUC-1, CEA, erbB2 and CA19-9 antigens were also detected in a significant number of tumor samples from GBC patients (Fig. 1d), suggesting that these were suitable antigen targets for immunotherapy approaches.

Heat shock induces the production of DAMPs in GBCCL

For the last 15 years, we have been developing a DC-based immunotherapy that improves the long-term survival of patients with advanced melanoma [11]. In our approach, a lysate derived from a mix of three heat shock-conditioned allogeneic melanoma cells (Mel1, Mel2, and Mel3), named TRIMEL, has been used as a source of both TAAs and DAMPs. Heat shock-induced DAMPs, particularly plasma

1901Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

membrane translocated eCRT and released HMGB1, medi-ate an optimal antigen presenting cell (APC) maturation and antigen cross-presentation, providing a unique strategy to obtain efficient tumor antigen-presenting cells with a mature DC-like phenotype [13].

Here, we evaluated the production of three common DAMPs (released HMGB1 and ATP, and translocated eCRT) in GBCCL subjected to heat shock. Heat shock treatment induced HMGB1 and ATP release in four of the eight cell lines evaluated (14TKB, G415, GBd1 and NOZ for ATP; and 2TKB, 24TKB, G415 and OCUG1 for HMGB1) (Fig. 2a, b). Additionally, three GBCCL translocated eCRT to the plasma membrane in response to heat shock (2TKB, GBd1 and OCUG1) (Fig. 2c, d). The levels of heat shock-induced DAMPs in GBCCL were similar that those induced

in the melanoma cell lines Mel1, Mel2 and Mel3, which were used as positive controls.

Heat shock‑conditioned GBCCL lysate mixtures, but not lysates from individual cell lines, induce differentiation of activated monocytes into mature DCs

As previously reported [13, 20], the addition of TRIMEL to IL-4/GM-CSF-activated monocytes (AM) mediated up to threefold induction of surface markers associated with DC maturation such as HLA-DR, CD80 and CD86 (Fig. 3a). However, heat shock-conditioned lysates prepared from each of the GBCCL did not induce a significant increase in the expression of these markers in stimulated AM (Fig. 3a).

Fig. 1 Tumor associated antigen expression in GBCCL and GBC fresh tumor samples. a Representative histograms for CA19-9, MUC-1, CEA, erbB2 and survivin expression in GBCCL evaluated by flow cytometry. Grey histograms indicate isotype control staining. b mRNA expression profiles for MAGE 1, 2, 3, GAGE 1/2 and BAGE in the GBCCL analyzed by RT-PCR. Actin was used as a housekeep-

ing gene control. c Summary of tumor associated antigen expression in GBCCL. Green and red refers to positive or negative expression, respectively. ND not determined. d Representative photomicrographs of immunohistochemical staining for CA19-9, MUC-1, CEA, erbB2 and survivin in paraffin-embedded tumor biopsies obtained from Chilean GBC patients (scale bar, 40 µm)

1902 Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

Given that the combination of different molecular factors present in each cell line may synergistically contribute to the DC stimulatory activity of the conditioned lysates as in TRI-MEL, we produced eight different heat shock-conditioned lysates (M1-M8) combining three different GBCCL in each lysate. The cell lines composing each mixture lysate are described in the Methods section and were chosen accord-ing to their tumor antigen expression and presence of heat shock-inducible DAMPs. Unlike individual cell lysates, GBCCL mixture lysates significantly induced the expres-sion of CD80, CD86 and HLA-DR in DCs (Fig. 3b). We extended the analysis to three additional markers: HLA-ABC, CD83 and CCR7 for four of these mixtures of GBCCL lysates: M2, M3, M5 and M8 (Fig. 3c), which were selected considering the antigen expression and DAMP production of the composing cells and the DC stimulatory activity of the lysate. The addition of M2, M3, M5, M8 or TRIMEL lysates mediated the induction of these maturation markers in DCs (Fig. 3c).

DCs matured with GBCCL lysates induced the activation of allogeneic CD4+ and  CD8+ T cells

To determine GBCCL lysates with major potential to induce mature DCs, we investigated the capacity of DCs matured

with the GBCCL lysates M2, M3, M5 and M8 (named M2-DCs, M3-DCs, M5-DCs and M8-DCs, respectively) or with TRIMEL (TRIMEL-DCs, as a positive control) to activate allogeneic T cells. After 5 days of DC/T cell co-cultures, we evaluated the surface expression of the lympho-cyte activation markers CD25 and CD69 and the chemokine receptors CXCR3 and CXCR4 on CD4+ and CD8+ T cells. All the DCs tested induced increased levels of CD25 and CD69 in both subsets (Fig. 4a). Moreover, all DCs were able to induce the expression of both receptors CXCR3 and CXCR4 in CD4+ T cells (Fig. 4a) whereas only the chemokine receptor CXCR3 was induced in CD8+ T cells co-cultured with all the DC types (Fig. 4a). Additionally, our results demonstrated that both CD4+ and CD8+ T cells co-cultured with allogeneic DCs loaded with GBCCL heat shock-conditioned lysates expressed high levels of the Th1 cytokines IFN-γ and TNF-α, whereas co-cultured CD8+ but not CD4+ T cells expressed the Th2 polarizing cytokine IL-4 (Fig. 4b–d). Finally, all the DCs evaluated induced the proliferation of both CD4+ and CD8+ allogeneic T cells (Fig. 4e).

Based on these results, we selected M2-DCs (loaded with heat shock-conditioned lysate from 2TKB, 24TKB and GBd1 GBCCL) for further experiments. The cell lines com-posing the M2 lysate were adenocarcinoma cell lines (the

Fig. 2 Heat shock conditioning induces DAMP production in GBCCL. The levels of ATP (a) or HMGB1 (b) were evaluated in the supernatants from heat shock-treated or control cells. c Representa-tive histograms showing the extracellular expression levels of trans-located calreticulin (eCRT) in heat shock-treated (dark grey) or con-trol (light grey) melanoma and GBC cells. White histograms indicate

isotype control staining. The percentage of eCRT positive (eCRT pos) for each condition is shown. d Statistical analysis of eCRT transloca-tion induced by heat shock in GBCCL. Bars represent averages and standard deviations of three (b–d) or five-seven (a) measurements of three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001

1903Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

most common histology of GBC) and combined they could provide a complete panel of TAAs and DAMPs (Figs. 1c, 2).

T cells activated by autologous M2‑DCs recognize HLA‑A2‑matched GBCCL

Given that the heat shock-conditioned M2 lysate poten-tially contains a large number of GBC tumor-antigenic epitopes for priming T cell responses, we investigated whether CD8+ tumor-specific IFN-γ-secreting T cells were also being elicited in vitro by autologous HLA-A2+ M2-DCs. First we observed that M2-DCs were able to activate autologous CD4+ and CD8+ T cells, measured by the percentage of T cells expressing CD25 and CD69 after 14 days of co-culture (Fig. 5a, b). Then, CD8+ T cells were isolated after co-culture by cell-sorting and

challenged with two HLA-A2+ GBCCL present in the M2 lysate (2TKB and GBd1), a HLA-A2+ GBCCL that was not included in the M2 lysate (CAVE), a HLA-A2+ mela-noma cell line (Mel1), or with K562 cells as a negative control. After challenging with 2TKB, GBd1 or CAVE cells, M2-DC-activated CD8+ T cells released signifi-cantly higher levels of IFN-γ than CD8+ T cells unstimu-lated or co-cultured with AM or TRIMEL-DCs (Fig. 5c). The NK cell-sensitive cell line K562 did not induce IFN-γ release by the activated CD8+ T cells. Additionally, we observed that there was an important cross-recognition of melanoma cells by T cells activated with M2-DCs (Fig. 5c). Similarly, T cells activated with TRIMEL-DCs were able to cross-recognize GBC cells, which may be indicative of shared antigens between both kinds of tumor cells.

Fig. 3 Heat shock-conditioned GBCCL lysate mixtures, but not lysates from individual cell lines, induce differentiation of activated monocytes into mature DCs. Surface expression of HLA-DR, CD80, CD86 (a, b), and HLA-ABC, CD83, and CCR7 (c) were evaluated by flow cytometry on activated monocytes (AM) incubated or not for 24 h with 100 µg/mL of heat shock-conditioned tumor lysates gener-ated from individual GBCCL (a) or mixtures (M1-M8) of three dif-ferent GBCCL (b, c). Bars represent the average and SD of the fold

induction of the integrated MFI (iMFI: % positive cells × geoMFI of positive cells) for each marker relative to AM from at least three inde-pendent experiments. Evaluated cell lysates mix were made as fol-lows: M1 (24TKB + GBd1 + G415); M2 (2TKB + 24TKB + GBd1); M3 (1TKB + 2TKB + 24TKB); M4 (OCUG1 + GBd1 + G415); M5 (2TKB + G415 + OCUG1); M6 (NOZ + OCUG 1 + G415); M7 (1TKB + 14TKB + 24TKB); and M8 (24TKB + OCUG1 + G415). *p < 0.05; **p < 0.01; ***p < 0.001

1904 Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

Discussion

Exploration of new active immunotherapies as complements to the relatively recent approaches grounded on blockade of immune checkpoint molecules, such as cytotoxic T-lym-phocyte antigen 4 (CTLA-4), programmed death (PD)-1 and PD-ligand-1 (PD-L1), may constitute a feasible possibility for improvement of clinical response rates. Particularly, DC-based cancer vaccines again become an interesting alterna-tive because of their relative effectiveness in activating cell-mediated immune responses and lack of severe side effects in patients [21]. In this context, whole tumor cell lysates are excellent sources for the delivery of a wide range of TAAs

that will generate MHC class I/II T cell epitopes for induc-ing the activation of CD4+ T helper and CD8+ cytotoxic T cells simultaneously, and therefore, a more integral immune response.

One method to determine the potential usefulness of DC-based immunotherapy in GBC patients is to explore the immunogenicity of GBC tumors by measuring the impact of T cell subpopulation infiltration at tumor sites and to cor-relate this with the overall survival of patients. Tumor-infil-trating immune cells constitute an accepted manifestation of the host immune response against cancer. Likewise, a rela-tionship between tumor-infiltrating immune cells and GBC prognosis has been suggested. In fact, recent published data

Fig. 4 Activation of allogeneic T cells by monocyte-derived DCs matured with different heat shock-conditioned GBC lysates. Purified CD3+ T cells were co-cultured for 5 days with allogeneic TRIMEL-, M2-, M3-, M5-, M8-DCs or without DCs. The surface expression of CD25, CD69, CXCR3 and CXCR4 (a), the intracellular levels of IFN-γ, TNF-α and IL-4 (b–d), and proliferation (e) were evaluated in the CD4+ and CD8+ T cells populations by flow cytometry. a, d Bars represent the average and SD from five independent experi-ments of the % of T cells positive for each marker, with the excep-

tion of CXCR3 and CXCR4 data that are shown as fold induction of the MFI relative to unstimulated T cells. Representative dot plots of IFN-γ and TNF-α production in allogeneic CD4+ (b) and CD8+ (c) T cells co-cultured with M2-DCs. e The percentage and SD of prolif-erating T cells are showed on the left of each histograms. Evaluated cell lysates mix were made as follows: M2 (2TKB + 24TKB + GBd1); M3 (1TKB + 2TKB + 24TKB); M5 (2TKB + G415 + OCUG1); and M8 (24TKB + OCUG1 + G415). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 (comparison versus unstimulated T cells)

1905Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

from our and other groups showed that CD8+ T cell infil-tration at different disease stages correlates with improved survival of GBC patients [22–24]. In one study, in which 45 tumor samples from GBC patients and 65 benign gall-bladder tissues were examined, increased frequencies of CD4+, CD8+ T cells and DCs were observed in GBC sam-ples, which significantly correlated with prolonged patient survival [23]. In a more recent study, Oguro and cowork-ers [25] analyzed 211 GBC samples and found that a lower density of tumor-infiltrating CD8+ cells and higher ratios between Foxp3+/CD4+, B and T lymphocyte attenuator/

CD8+, and casitas-B-lineage lymphoma protein-b/CD8+ were significantly associated with shorter overall survival in GBC patients. Moreover, in a cohort of 80 Chilean GBC patients, we observed that a greater infiltration of CD8+ T cells in cancer tissue was associated with a favorable prog-nostic biomarker for both early and advanced stage patients [24]. Altogether, these observations strongly indicate that a natural host CD8+ T cell-mediated immune response against GBC increases patient survival. These findings encourage the design and development of adjuvant immunotherapeutic approaches against GBC.

Fig. 5 T cells activated by autologous monocyte-derived DCs loaded with a heat shock conditioned GBC lysate recognize HLA-A2-matched GBCCL. a–c Purified CD3+ T cells were co-cultured for 14  days with autologous HLA-A2+ AM, TRIMEL-DCs, M2-DCs or cultured alone. The surface expression of CD25, CD69, CXCR3 and CXCR4 (a, b) were evaluated in the CD4+ (a) and CD8+ (b) T cells populations by flow cytometry. Bars represent the average and SD from at least three independent experiments of the % of T cells positive for each marker, with the exception of CXCR3 and CXCR4 data that are shown as fold induction of the MFI relative to unstimu-

lated T cells. *p < 0.05; **p < 0.01; ***p < 0.001 (comparison versus unstimulated T cells). c Sorted CD8+ T cells were challenged for 16 h with the HLA-A2+ GBCCL 2TKB, GBd1, CAVE, the melanoma cell line Mel1 or K562 cells. IFN-γ release was measured by ELIS-POT at different effector:target ratios as indicated. Data represent the average and SD of at least three independent experiments. *p < 0.05; ***p < 0.001; ****p < 0.0001 (comparison M2-DC versus TRIMEL-DCs stimulated T cells). M2 refer to the mixture made from three dif-ferent GBCCL

1906 Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

Tabl

e 1

Cur

rent

pro

toco

ls u

sed

for w

hole

tum

or ly

sate

pre

para

tions

in im

mun

ogen

ic D

C v

acci

nes

Stra

tegy

TAA

s/ne

o-A

gsW

hole

tum

or c

ells

Who

le tu

mor

cel

ls +

cyto

kine

sTu

mor

lysa

tes

Con

ditio

ned

tum

or ly

sate

s

Mec

hani

sm

Isol

ated

Ags

Tum

or

Mol

ecul

ar

Cha

ract

eriz

atio

n

Irra

diat

e

Aut

olog

ous

or

allo

gene

ic

GM

-CS

F

Irra

diat

ed

Tra

nsdu

ced

or

com

bine

d

Lysis

TAAs

UV

Hea

t S

ho

ck

Lysis

TAA

DAMPs

HOC

Adv

anta

ges

Spec

ific

imm

uno-

dom

inan

t A

gsN

eoA

gs-r

eact

ive

T ce

lls a

re

not a

ffect

ed b

y ce

ntra

l tol

er-

ance

Pers

onal

ized

Bro

ad ra

nge

of T

AA

sH

aplo

type

-inde

pend

ent

Pers

onal

ized

repe

rtoire

of

TAA

s (au

tolo

gous

)

Bro

ad ra

nge

of T

AA

sH

aplo

type

-inde

pend

ent

Cyt

okin

es in

duce

recr

uitm

ent,

activ

atio

n or

pro

lifer

atio

n of

im

mun

e ce

lls

Bro

ad ra

nge

of T

AA

sH

aplo

type

-inde

pend

ent

Stan

dard

ized

pre

para

tion

and

feas

ibili

ty o

f mas

s pro

duc-

tion

Bro

ad ra

nge

of T

AA

sH

aplo

type

-inde

pend

ent

Pres

ence

of m

atur

atio

n sti

mul

i fo

r DC

sFe

asib

ility

of m

ass a

nd st

anda

rd

prod

uctio

nD

isad

vant

ages

Lack

of A

g di

vers

ity a

nd D

C

mat

urat

ion

stim

uli

Imm

une

evas

ion

of tu

mor

ce

lls la

ckin

g sp

ecifi

c-A

g ex

pres

sion

Lack

of D

C m

atur

atio

n sti

mul

iG

ener

atio

n of

imm

uno-

sup-

pres

sive

mol

ecul

esLi

mite

d to

pat

ient

s with

su

rgic

ally

acc

essi

ble

tum

ors

(aut

olog

ous)

Lack

of p

atie

nt-s

peci

fic A

gs

(allo

gene

ic)

Lack

of D

C m

atur

atio

n sti

mul

iIr

radi

atio

n co

uld

gene

r-at

e im

mun

o-su

ppre

ssiv

e m

olec

ules

Lack

of p

atie

nt-s

peci

fic A

gs

(allo

gene

ic)

Lack

of D

Cs m

atur

atio

n sti

mul

iPo

ssib

le p

rese

nce

of im

mu-

nore

gula

tory

mol

ecul

es

from

tum

or c

ells

(i.e

., IL

-10,

TG

F-β)

Lack

of p

atie

nt-s

peci

fic A

gs

(allo

gene

ic)

Poss

ible

pre

senc

e of

imm

u-no

regu

lato

ry m

olec

ules

from

th

e tu

mor

cel

ls (i

.e.,

IL-1

0,

TGF-

β)La

ck o

f pat

ient

-spe

cific

Ags

(a

lloge

neic

)

Clin

ical

Out

com

es (p

atie

nts s

tage

; tri

al´s

pha

se; t

ype

of

tum

or)

Incr

ease

d ne

oAg-

spec

ific

T ce

lls a

nd d

etec

tion

of H

LA

clas

s I-r

estri

cted

neo

-Ags

(I

II; p

hase

I; m

elan

oma)

[32]

4/6

patie

nts w

ithou

t rec

ur-

renc

e af

ter 2

5 m

onth

s pos

t-va

ccin

atio

n (I

II/IV

; pha

se I;

m

elan

oma)

[33]

Ado

ptiv

e ce

ll tra

nsfe

r of

mut

atio

n-sp

ecifi

c Th

1 ce

lls

led

to tu

mor

regr

essi

on a

nd

stab

iliza

tion

of d

isea

se (I

V;

phas

e I;

chol

angi

ocar

cino

ma)

[3

4]

Incr

ease

d ov

eral

l sur

viva

l (O

S)

(39%

vs 2

0%) (

IV; p

hase

II;

mel

anom

a) [3

5]In

crea

sed

med

ian

dise

ase

free

su

rviv

al (M

DFS

) fro

m 7

to

20

mon

ths (

III;

phas

e II

; m

elan

oma)

[36]

Mea

n O

S of

21.

9 m

onth

s. Po

sitiv

e co

rrel

atio

n be

twee

n an

tibod

y tit

er a

nd O

S (I

V;

phas

e II

; col

on) [

37]

Patie

nts d

evel

oped

imm

une

reac

tions

aga

inst

the

tum

or.

One

pat

ient

with

com

plet

e tu

mor

regr

essi

on (I

II; p

hase

I;

ovar

ian)

[38]

Mea

n di

seas

e fr

ee in

terv

al o

f 28

.8 m

onth

s (IV

; pha

se I;

br

east)

[39]

Act

ivat

ion

of th

e im

mun

e sy

s-te

m b

ut n

o tu

mor

regr

essi

on

(III

/IV; p

hase

I/II

; lun

g) [4

0]M

edia

n O

S of

34.

9 m

onth

s (h

igh

dose

gro

up) c

ompa

red

to 2

4 m

onth

s (lo

w d

ose

grou

p) (I

V; p

hase

I/II

; pro

s-ta

te) [

41]

Low

toxi

city

and

gen

erat

ion

of c

ellu

lar i

mm

une

resp

onse

(I

I/III

/IV; p

hase

I; m

ela-

nom

a) [4

2]50

% o

f pat

ient

s with

stab

le

dise

ase

and

a m

edia

n O

S of

40

mon

ths c

ompa

red

with

hi

storic

med

ian

OS

of 1

0 m

onth

s (IV

; Pha

se I;

rena

l) [4

3]

Med

ian

OS

of 1

1.5 

mon

ths

(com

pare

d w

ith h

istor

ic

med

ian

OS

of 9

.6 m

onth

s)

and

1- a

nd 2

-yea

r sur

viva

l ra

tes o

f 50

and

27%

, res

pec-

tivel

y. 7

/22

patie

nts w

ith

stab

le d

isea

se (I

V; p

hase

II;

mes

othe

liom

a) [4

4]Su

rviv

al b

enefi

t in

patie

nts

expr

essi

ng th

e A

gs H

LA-A

2 an

d H

LA-C

3 (3

8% o

f clin

i-ca

l res

pons

e ra

te v

ersu

s 7%

in

pat

ient

s with

out e

xpre

s-si

on o

f the

Ags

) (II

I/IV

; ph

ase

II/II

I; m

elan

oma)

[45]

2/14

pat

ient

s exp

erie

nced

m

inor

or p

artia

l res

pons

es

(tum

or si

ze d

ecre

ase)

(IV

; ph

ase

I; m

elan

oma)

[46]

2 ou

t of 5

pat

ient

s exp

erie

nced

pr

ogre

ssio

n-fr

ee su

rviv

al

inte

rval

s of 3

6 an

d 44

mon

ths

(II/I

V; p

hase

I; o

varia

n) [4

7]St

age

IV D

TH+

pat

ient

s sh

owed

a m

edia

n O

S of

33

mon

ths c

ompa

red

with

the

11

mon

ths f

rom

DTH

− p

atie

nts.

All

stag

e II

I pat

ient

s wer

e D

TH+

and

rem

aine

d tu

mor

-fr

ee fo

r a m

edia

n fo

llow

-up

of

48 m

onth

s (II

I/IV

; pha

se II

; m

elan

oma)

[11]

1907Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

The aforementioned GBC T cell infiltration might be orchestrated by the chemokine receptor CXCR4, given that its ligand, C-X-C motif ligand-12 (CXCL12), is fre-quently overexpressed in GBC [26]. Likewise, the expres-sion of CXCR3 by lymphocytes can mediate its migration to GBC tumor beds [27]. These data suggest that the induc-tion of these chemokine receptors in T cells by therapeutic DCs would be beneficial for the DC-mediated anti-tumor responses in vaccinated patients.

The potential use of immunotherapeutic approaches for GBC has only recently become a subject of intensive inves-tigation. In fact, current immunotherapies against GBC have been focused on the use of peptide-based vaccines or peptide-loaded DCs [21, 28]. These strategies have shown modest clinical improvements, likely due to induced toler-ance by dominant single tumor peptides or by the selection of antigen loss variants in established tumors. In contrast, a study where DC loaded with autologous tumor cell lysates combined with activated T cell transfer were used as an adju-vant treatment in operated patients with advanced intrahe-patic cholangiocarcinoma, reported improved post-operative progression-free and overall survival compared to patients receiving surgery alone [29].

The optimal delivery of tumor antigens is one of the most important factors for the success of DC-based anti-cancer vaccines. With this in mind, lysates from allogeneic tumor cells, whole tumor cells, tumor mRNA, and antigenic pep-tides, have all been tested as tumor vaccines. Autologous whole tumor antigens offer an unparalleled advantage as it allows DCs to process and present a broad range of TAAs to stimulate strong, polyclonal and long-term memory CD4+ and CD8+ T cell responses, potentially preventing tumor immune escape. Moreover, this strategy is suitable for all cancer patients regardless of their HLA haplotype. However, not all cancer patients have surgically removable tumors, and therefore, a useful and promising alternative is the preparation of allogeneic cancer cell lysates that have demonstrated to provide a standardized applicable source of tumor-specific antigens in patients with non-resectable tumors [30]. Importantly, the method used for inducing cell death or protein chemical modifications during whole tumor lysate preparation could impact the immunogenicity and effi-cacy of the therapy (Table 1). Current immunogenic treat-ment modalities used for pre-conditioning tumor cell lysates include ultraviolet irradiation, oxidation-inducing modalities and heat shock treatments [31]. In the present study, we gen-erated heat shock-conditioned tumor lysate for GBC (M2), which have some important characteristics that suggest its potential as an antigen source for DC vaccines: (1) it con-tains a broad panel of TAAs, also expressed in tumors from GBC patients, (2) it includes different molecules that could act as DAMPs (released HMGB1, ATP and eCRT), (3) it promotes a rapid and efficient differentiation of monocytes

to mature DCs, and (4) DCs generated with this lysate are able to induce the activation of T cells that specifically rec-ognize tumor cells.

In general, in vivo tumor antigen presentation by immu-notherapeutic DCs might drive the development of tumor-specific adaptive immune responses, whereas cytotoxic CD8+ T cells recognize and attack tumor cells through rec-ognition of TAA peptides associated to MHC class I. There-fore, T-cell cytotoxicity depends on MHC class I expression on tumor cell surface. It has been frequently observed that tumor cells lost MHC class I expression, and therefore, the efficacy of DC-mediated immunotherapies may be reduced. In line with this hypothesis, it has been shown that reduced MHC class I expression in biliary tract cancers, including GBC, was linked to shortened overall patient survival [48]. However, in the majority of cases the loss of MHC class I is partial, affecting only some isotypes, and thus an important portion of cancer patients could benefit from DC-mediated immunotherapy. Moreover, it is very important to incorpo-rate strategies to recover MHC class I expression in tumors to improve immunotherapy effect [49]. In conclusion, we propose that GBC cell lysate-loaded DCs may be considered for future immunotherapy approaches alone or in combi-nation with currently used immune checkpoint molecule-blocking therapies.

Acknowledgements We thank Marisol Briones for administrative and technical support, and Benedict Chambers for critical reading and Eng-lish editing of the manuscript.

Author contributions DR-S contributed to the experimental design and conducted most of the experiments shown in this paper; IG and JMB contributed to the collection of tumor biopsies; JCR contributed to the establishment of the different GBCCLs; DR-S, IG and AT contributed to the statistical analysis conducted in this paper; FS-O designed and coordinated the study; DR-S, AT, MAG, IÁ, CP, FEG, PF, MNL and FS-O analyzed the data and wrote the manuscript. All authors had full access to all reported data and take responsibility for the integrity of the data and the accuracy of the data analysis. All authors take responsibil-ity for the preparation of the manuscript.

Funding This work was supported by grants from the National Fund for Scientific and Technological Development (FONDECYT) 1171213 and 11160380; the Fund for the Promotion of Scientific and Techno-logical Development (FONDEF) ID16i10148; Millennium Institute on Immunology and Immunotherapy (MIII) P09/016-F; National Commis-sion for Scientific and Technological Research (CONICYT)-PCHA/Doctorado Nacional/21130465; CONICYT-PAI/TESIS DOCTORADO SECTOR PRODUCTIVO/7815110008; FONDECYT POSTDOCT-ORADO 3170917.

Compliance with ethical standards

Conflict of interest The authors declare that they have no conflict of interest.

Ethical approval and ethical standards The use of paraffin embedded GBC tissue of the patients was approved by the bioethical committee

1908 Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

of the Hospital Clínico of the Universidad de Chile (Approval act 75), which authorize the use of biopsies under the supervision of the direc-tor of the Pathological Anatomy Service (Dr. Iván Gallegos). Informed consent from patients was not required for this work, given that it corre-sponds to a retrospective study. The Bioethical Committee of the Cen-tro Metropolitano de Sangre y Tejidos, Hospital Metropolitano (San-tiago de Chile) approved the use of buffy coats from healthy donors.

Open Access This article is distributed under the terms of the Crea-tive Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribu-tion, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

References

1. Rojas-Sepúlveda D, Gleisner MA, Pereda C, López MN, Salazar-Onfray F (2016) Tumor cell lysates as maturation stimulus and antigen source for therapeutic dendritic cells against gallbladder cancer. Eur J Immunol 46(S1):1221–1222 (poster)

2. Rojas-Sepúlveda D, Gleisner MA, Pereda C, López MN, Sala-zar-Onfray F (2017) Tumor lysate loaded dendritic cells induce a T cell specific antitumor response against gallbladder cancer. J Immunol 198(1 Supplement):79.18 (poster)

3. Guida M, Pisconte S, Colucci G (2012) Metastatic melanoma: the new era of targeted therapy. Expert Opin Ther Targets 16 Suppl 2:S61–S70. https ://doi.org/10.1517/14728 222.2011.64580 7

4. Larkin J, Chiarion-Sileni V, Gonzalez R, Grob JJ, Cowey CL, Lao CD et al (2015) Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med 373:23–34. https ://doi.org/10.1056/NEJMo a1504 030

5. Farolfi A, Ridolfi L, Guidoboni M, Nicoletti SV, Piciucchi S, Val-morri L et al (2012) Ipilimumab in advanced melanoma: reports of long-lasting responses. Melanoma Res 22:263–270. https ://doi.org/10.1097/CMR.0b013 e3283 53e65 c

6. Brahmer JR, Drake CG, Wollner I, Powderly JD, Picus J, Sharfman WH et al (2010) Phase I study of single-agent anti-programmed death-1 (MDX-1106) in refractory solid tumors: safety, clinical activity, pharmacodynamics, and immunologic correlates. J Clin Oncol 28:3167–3175. https ://doi.org/10.1200/JCO.2009.26.7609

7. Simeone E, Ascierto PA (2012) Immunomodulating antibodies in the treatment of metastatic melanoma: the experience with anti-CTLA-4, anti-CD137, and anti-PD1. J Immunotoxicol 9:241–247. https ://doi.org/10.3109/15476 91X.2012.67802 1

8. Hamid O, Robert C, Daud A, Hodi FS, Hwu WJ, Kefford R et al (2013) Safety and tumor responses with lambrolizumab (anti-PD-1) in melanoma. N Engl J Med 369:134–144. https ://doi.org/10.1056/NEJMo a1305 133

9. Kyi C, Postow MA (2016) Immune checkpoint inhibitor combi-nations in solid tumors: opportunities and challenges. Immuno-therapy 8(7):821–837. https ://doi.org/10.2217/imt-2016-0002

10. Andrews DM, Maraskovsky E, Smyth MJ (2008) Cancer vaccines for established cancer: how to make them better? Immunol Rev 222:242–255. https ://doi.org/10.1111/j.1600-065X.2008.00612 .x

11. López MN, Pereda C, Segal G, Muñoz L, Aguilera R, González FE et al (2009) Prolonged survival of dendritic cell-vaccinated melanoma patients correlates with tumor-specific delayed type IV hypersensitivity response and reduction of tumor growth factor beta-expressing T cells. J Clin Oncol 27(6):945–952. https ://doi.org/10.1200/JCO.2008.18.0794

12. González FE, Gleisner A, Falcón-Beas F, Osorio F, López MN, Salazar-Onfray F (2014) Tumor cell lysates as immunogenic sources for cancer vaccine design. Hum Vaccin Immunother 10(11):3261–3269. https ://doi.org/10.4161/21645 515.2014.98299 6

13. Aguilera R, Saffie C, Tittarelli A, González FE, Ramírez M, Reyes D et al (2011) Heat-shock induction of tumor-derived danger sig-nals mediates rapid monocyte differentiation into clinically effec-tive dendritic cells. Clin Cancer Res 17(8):2474–2483. https ://doi.org/10.1158/1078-0432.CCR-10-2384

14. Hundal R, Shaffer EA (2014) Gallbladder cancer: epidemiology and outcome. Clin Epidemiol 6:99–109. https ://doi.org/10.2147/CLEP.S3735 7

15. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, Parkin DM, Forman D, Bray F (2015) Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 136(5):E359–E386. https ://doi.org/10.1002/ijc.29210

16. Roa I, Ibacache G, Muñoz S, de Aretxabala X (2014) Gallblad-der cancer in Chile: pathologic characteristics of survival and prognostic factors: analysis of 1,366 cases. Am J Clin Pathol 141(5):675–682. https ://doi.org/10.1309/AJCPQ T3ELN 2BBCK A

17. Roa I, de Aretxabala X (2015) Gallbladder cancer in Chile: what have we learned? Curr Opin Gastroenterol 31(3):269–275. https ://doi.org/10.1097/MOG.00000 00000 00016 4

18. Smok G, Cervilla K, Bosch H, Csendes A (1986) Precancerous lesions of invasive carcinoma of the gallbladder. Rev Med Chil 114(10):954–958

19. Sheth S, Bedford A, Chopra S (2000) Primary gallbladder cancer: recognition of risk factors and the role of prophylactic cholecys-tectomy. Am J Gastroenterol 95(6):1402–1410. https ://doi.org/10.1111/j.1572-0241.2000.02070 .x

20. Escobar A, López M, Serrano A, Ramirez M, Pérez C, Aguirre A, González R, Alfaro J, Larrondo M, Fodor M, Ferrada C, Salazar-Onfray F (2005) Dendritic cell immunizations alone or combined with low doses of interleukin-2 induce specific immune responses in melanoma patients. Clin Exp Immunol 142(3):555–568. https ://doi.org/10.1111/j.1365-2249.2005.02948 .x

21. Marks EI, Yee NS (2016) Molecular genetics and targeted therapeutics in biliary tract carcinoma. World J Gastroenterol 22(4):1335–1347. https ://doi.org/10.3748/wjg.v22.i4.1335

22. Goeppert B, Frauenschuh L, Zucknick M, Stenzinger A, Andru-lis M, Klauschen F et al (2013) Prognostic impact of tumour-infiltrating immune cells on biliary tract cancer. Br J Cancer 109(10):2665–2674. https ://doi.org/10.1038/bjc.2013.610

23. Nakakubo Y, Miyamoto M, Cho Y, Hida Y, Oshikiri T, Suzuoki M, Hiraoka K, Itoh T, Kondo S, Katoh H (2003) Clinical signifi-cance of immune cell infiltration within gallbladder cancer. Br J Cancer 89(9):1736–1742. https ://doi.org/10.1038/sj.bjc.66013 31

24. Fluxá P, Rojas-Sepúlveda D, Gleisner MA, Tittarelli A, Villegas P, Tapia L et al (2018) High CD8(+) and absence of Foxp3(+) T lymphocytes infiltration in gallbladder tumors correlate with prolonged patients survival. BMC Cancer 18(1):243. https ://doi.org/10.1186/s1288 5-018-4147-6

25. Oguro S, Ino Y, Shimada K, Hatanaka Y, Matsuno Y, Esaki M et al (2015) Clinical significance of tumor-infiltrating immune cells focusing on BTLA and Cbl-b in patients with gallbladder cancer. Cancer Sci 106(12):1750–1760. https ://doi.org/10.1111/cas.12825

26. Lee HJ, Lee K, Lee DG, Bae KH, Kim JS, Liang ZL et al (2012) Chemokine (C-X-C motif) ligand 12 is associated with gallblad-der carcinoma progression and is a novel independent poor prog-nostic factor. Clin Cancer Res 18(12):3270–3280. https ://doi.org/10.1158/1078-0432.CCR-11-2417

27. Patil RS, Shah SU, Shrikhande SV, Goel M, Dikshit RP, Chiplunkar SV (2016) IL17 producing γδT cells induce

1909Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

angiogenesis and are associated with poor survival in gallblad-der cancer patients. Int J Cancer 139(4):869–881. https ://doi.org/10.1002/ijc.30134

28. Chai Y (2016) Immunotherapy of biliary tract cancer. Tumour Biol 37(3):2817–2821. https ://doi.org/10.1007/s1327 7-015-4743-x

29. Shimizu K, Kotera Y, Aruga A, Takeshita N, Takasaki K, Yama-moto M (2012) Clinical utilization of postoperative dendritic cell vaccine plus activated T-cell transfer in patients with intrahepatic cholangiocarcinoma. J Hepatobiliary Pancreat Sci 19(2):171–178. https ://doi.org/10.1007/s0053 4-011-0437-y

30. Chiang CL, Coukos G, Kandalaft LE (2015) Whole tumor antigen vaccines: where are we? Vaccines (Basel) 3(2):344–372. https ://doi.org/10.3390/vacci nes30 20344

31. Vandenberk L, Belmans J, Van Woensel M, Riva M, Van Gool SW (2016) Exploiting the immunogenic potential of cancer cells for improved dendritic cell vaccines. Front Immunol. https ://doi.org/10.3389/fimmu .2015.00663

32. Carreno BM, Magrini V, Becker-Hapak M, Kaabinejadian S, Hundal J, Petti AA et al (2015) Cancer immunotherapy. A den-dritic cell vaccine increases the breadth and diversity of melanoma neoantigen-specific T cells. Science 348(6236):803–808. https ://doi.org/10.1126/scien ce.aaa38 28

33. Ott PA, Hu Z, Keskin DB, Shukla SA, Sun J, Bozym DJ et al (2017) An immunogenic personal neoantigen vaccine for patients with melanoma. Nature 547(7662):217–221. https ://doi.org/10.1038/natur e2299 1

34. Tran E, Turcotte S, Gros A, Robbins PF, Lu YC, Dudley ME, Parkhurst MR (2014) Cancer immunotherapy based on mutation-specific CD4+ T cells in a patient with epithelial cancer. Science 344(6184):641–645. https ://doi.org/10.1126/scien ce.12511 02

35. Hsueh EC, Essner R, Foshag LJ, Ollila DW, Gammon G, O’Day SJ, Boasberg PD, Stern SL, Ye X, Morton DL (2002) Pro-longed survival after complete resection of disseminated mela-noma and active immunotherapy with a therapeutic cancer vac-cine. J Clin Oncol 20(23):4549–4554. https ://doi.org/10.1200/JCO.2002.01.151

36. Mordoh J, Kairiyama C, Bover L, Solarolo E (1997) Allogeneic cells vaccine increases disease-free survival in stage III melanoma patients. A non randomized phase II study. Medicina (Buenos Aires) 57(4):421–427

37. Habal N, Gupta RK, Bilchik AJ, Yee R, Leopoldo Z, Ye W, Elashoff RM, Morton DL (2001) CancerVax, an allogeneic tumor cell vaccine, induces specific humoral and cellular immune responses in advanced colon cancer. Ann Surg Oncol 8(5):389–401

38. Berd D (2001) Autologous, hapten-modified vaccine as a treat-ment for human cancers. Vaccine 19(17):2565–2570. https ://doi.org/10.1016/S0264 -410X(00)00490 -4

39. Emens LA, Asquith JM, Leatherman JM, Kobrin BJ, Petrik S, Laiko M, Stearns V (2009) Timed sequential treatment with cyclophosphamide, doxorubicin, and an allogeneic granulocyte–macrophage colony-stimulating factor—secreting breast tumor

vaccine: a chemotherapy dose-ranging factorial study of safety and immune activation. J Clin Oncol 27(35):5911–5918. https ://doi.org/10.1200/JCO.2009.23.3494

40. Nemunaitis J, Jahan T, Ross H, Sterman D, Richards D, Fox B, Jablons D, Aimi J, Lin A, Hege K (2006) Phase 1/2 trial of autologous tumor mixed with an allogeneic GVAX vaccine in advanced-stage non-small-cell lung cancer. Cancer Gene Ther 13(6):555–562. https ://doi.org/10.1038/sj.cgt.77009 22

41. Small EJ, Sacks N, Nemunaitis J, Urba WJ, Dula E, Centeno AS et al (2007) Granulocyte macrophage colony-stimulating fac-tor—secreting allogeneic cellular immunotherapy for hormone-refractory prostate cancer. Clin Cancer Res 13(13):3883–3891. https ://doi.org/10.1158/1078-0432.CCR-06-2937

42. Barrio MM, de Motta PT, Kaplan J, von Euw EM, Bravo AI, Chacón RD, Mordoh J (2006) A phase I study of an allogeneic cell vaccine (VACCIMEL) with GM-CSF in melanoma patients. J Immunother 29(4):444–454. https ://doi.org/10.1097/01.cji.00002 08258 .79005 .5f

43. Westermann J, Flörcken A, Willimsky G, van Lessen A, Kopp J, Takvorian A et al (2011) Allogeneic gene-modified tumor cells (RCC-26/IL-7/CD80) as a vaccine in patients with metastatic renal cell cancer: a clinical phase-I study. Gene Ther 18(4):354–363. https ://doi.org/10.1038/gt.2010.143

44. Powell A, Creaney J, Broomfield S, Van Bruggen I, Robinson B (2006) Recombinant GM-CSF plus autologous tumor cells as a vaccine for patients with mesothelioma. Lung Cancer 52(2):189–197. https ://doi.org/10.1016/j.lungc an.2006.01.007

45. Sondak VK, Sosman JA (2003) Results of clinical trials with an allogeneic melanoma tumor cell lysate vaccine: Melacine. Semin Cancer Biol 13(6):409–415. https ://doi.org/10.1016/j.semca ncer.2003.09.004

46. Chang AE, Redman BG, Whitfield JR, Nickoloff BJ, Braun TM, Lee PP, Mulé JJ (2002) A phase I trial of tumor lysate-pulsed dendritic cells in the treatment of advanced cancer. Clin Cancer Res 8(4):1021–1032

47. Chiang CL, Kandalaft LE, Tanyi J, Hagemann AR, Motz GT, Svo-ronos N et al (2013) A dendritic cell vaccine pulsed with autolo-gous hypochlorous acid-oxidized ovarian cancer lysate primes effective broad antitumor immunity: from bench to bedside. Clin Cancer Res 19(17):4801–4815. https ://doi.org/10.1158/1078-0432.CCR-13-1185

48. Goeppert B, Frauenschuh L, Zucknick M, Roessler S, Mehrabi A, Hafezi M, Stenzinger A, Warth A, Pathil A, Renner M, Schir-macher P, Weichert W (2015) Major histocompatibility complex class I expression impacts on patient survival and type and density of immune cells in biliary tract cancer. Br J Cancer 113(9):1343–1349. https ://doi.org/10.1038/bjc.2015.337

49. Garrido F, Aptsiauri N, Doorduijn EM, Garcia Lora AM, van Hall T (2016) The urgent need to recover MHC class I in cancers for effective immunotherapy. Curr Opin Immunol 39:44–51. https ://doi.org/10.1016/j.coi.2015.12.007

Affiliations

Daniel Rojas‑Sepúlveda1,2,3 · Andrés Tittarelli1,2  · María Alejandra Gleisner1,2 · Ignacio Ávalos1,2 · Cristián Pereda1,2 · Iván Gallegos5 · Fermín Eduardo González2,4 · Mercedes Natalia López1,2 · Jean Michel Butte6 · Juan Carlos Roa7,8 · Paula Fluxá1,2 · Flavio Salazar‑Onfray1,2

1 Disciplinary Program of Immunology, Institute of Biomedical Sciences, Faculty of Medicine, Universidad de Chile, Independencia 1027, building H, Third floor, 8380453 Santiago, Chile

2 Millennium Institute on Immunology and Immunotherapy, Universidad de Chile, 8380453 Santiago, Chile

1910 Cancer Immunology, Immunotherapy (2018) 67:1897–1910

1 3

3 Faculty of Science, Universidad San Sebastián, Lota 2465, 7510157 Santiago, Chile

4 Department of Conservative Dentistry, Faculty of Dentistry, Universidad de Chile, 8380492 Santiago, Chile

5 Pathological Anatomy Service, Clinic Hospital, Universidad de Chile, 8380456 Santiago, Chile

6 Department of Surgery, Fundación Arturo López Pérez, Institute of Oncology, 7500921 Santiago, Chile

7 Department of Pathology, School of Medicine, Pontificia Universidad Católica de Chile, 8330023 Santiago, Chile

8 Center for Investigation in Translational Oncology (CITO), Advanced Center for Chronic Diseases (ACCDiS), School of Medicine, Pontificia Universidad Católica de Chile, 8330023 Santiago, Chile